This papers presents a low-cost electromyograph combined with marker-less pose detection using computer vision. The developed and build three channel electromyograph is tested by measuring the muscle activity of one leg, while the subject is performing squats. Simultaneously, a camera records the exercise and subsequently the image data is evaluated by OpenPose. We could show that this simple setup enables the user to evaluate the muscle activity of three independent muscles as function of the

more »

... nee angle. These results are in good agreement to the expected muscle activity. The sample-rate of the EMG device is 2 kHz. The overall cost of the developed device is under 100 €. To our knowledge, this is the first work combining these two methods for dynamic exercises. The method is well customizable for other sports due to the battery powered device and its handy size.